Summary

Internal obturator and/or semitendinosus muscle flaps were utilised for a novel

indication to reinforce primary appositional rectal wall repair in three dogs and one

cat in this case series. All three dogs in this study incurred rectal wall compromise

during surgical excision of anal sac tumours. The cat sustained bite wounds to the

perianal region resulting in abscessation and a rectal tear. Perioperative complications

and short-term outcome are described as detailed in the patient records. Additional

follow up was obtained by telephone interview with the referring veterinary surgeon

or the owner and the technique was successful in all cases. The application of an

internal obturator and/or semitendinosus muscle flap has the potential to reduce the

risk of rectal wall dehiscence after primary repair, and consequently the risk of

pararectal abscess or recto-cutaneous fistula formation.

Keywords

Internal obturator; semitendinosus; muscle flap; rectal wall.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

Word Count: 2738 including references in text

Introduction

The application of the internal obturator and semitendinosus muscle flaps for the

repair of perineal herniation in dogs and cats has been well documented in the

veterinary literature (Orsher & Johnston 1985, Orsher 1986, Chambers & Rawlings

1991, Welches et al. 1992, Mortari et al. 2005, Vnuk et al. 2005, Szabo et al. 2007,

Morello et al. 2015, Shaughnessy & Monet 2015). For this purpose, the main

advantage afforded by muscle transposition is structural support to the terminal

rectum, but it is also recognized that the additional vascular supply provided may

promote tissue healing and reduce the risk of dehiscence and recurrence of the hernia

(Orsher 1986, Aronson 2012). Use of the semitendinosus muscle flap has also been

described to improve faecal incontinence in one dog with an external anal sphincter

muscle defect; in this case the muscle flap was thought to have improved the passive

pressure of the terminal rectum by way of increased lateral and ventral structural

support (Doust & Sullivan 2003).

Caudal rectal wall injury can occur as a result of trauma, intentional oncological

resection or ischaemia resulting from dissection in the perineal region. The healing

potential of the caudal rectum is poor when compared to other areas of the

gastrointestinal tract, which can likely be explained by several factors: vascular

supply, bacterial load, distension by faecal boluses and lack of omentum (Hall et al.

1998, Kirby 2012, Williams 2012). In dogs, the main arterial supply to the rectum

arises from the cranial rectal artery with insignificant and variable contributions from

the middle and caudal rectal arteries (Goldsmid et al. 1993, Evans & de Lahunta

2013). The absence of vasa recti and multiple anastomoses between different vascular

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

networks means there is little collateral supply in case of individual vessel injury or

ligation, increasing the susceptibility of the rectal wall to ischaemic necrosis. In cats,

the middle and caudal rectal arteries appear to contribute more reliably to the terminal

rectum but anastomoses are still inferior when compared to the colonic and small

intestinal vasculature (Aronson 2012).

Failure of the caudal rectum to heal after injury results in para-rectal abscess and/or

recto-cutaneous fistula formation; this can be challenging to manage and requires

local decontamination of the affected soft tissues (with or without drain placement or

temporary colostomy) followed by successful closure of the rectal wall (Tobias 1994,

Aronson 2012, Skinner et al. 2016). Repair of the rectal wall using biomaterials such

as porcine small-intestinal submucosa, porcine dermal collagen or autologous fascia

lata has been reported (Frankland 1986, Stoll et al. 2002, Bongartz et al. 2005,

Schallberger et al. 2008, Aronson 2012). The efficacy of biomaterials for perineal

repair or wound reinforcement has been variable in previous studies. Heterologous

grafts pose a risk of immunological reactions, may not expedite healing, and may not

be commonly available (Frankland 1986, Stoll et al. 2002, Schallberger et al. 2008,

Aronson 2012). Whilst the risk of immunological reactions is avoided by autologous

graft use, lameness has been reported as a short-term consequence of fascia lata

harvesting and the surgical approach requires more extensive aseptic skin preparation

and intra-operative repositioning of the patient (Bongartz et al. 2005). Alternatively

caudal rectal injury may be addressed via rectal pull-through with anastomosis of the

rectum to the skin, though this has a high risk of complications including

incontinence, tenesmus, stricture, dehiscence and infection (Morello et al. 2008,

Nucci et al. 2014).

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

The aim of this paper is to report the use of internal obturator and semitendinosus

muscle transposition in the management of rectal wall injury in dogs and cats and to

describe patient outcomes.

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

Case Descriptions

Three dogs and one cat underwent internal obturator and/or semitendinosus muscle

transposition flaps at institution X for the management of existing or anticipated rectal

wall perforation or dehiscence. For both muscle transpositions, the patient was

positioned in sternal recumbency with the hindquarters elevated on a beanbag and the

hind limbs over the edge of the surgical table. The tail was pulled cranially over the

patient’s dorsum and tied under tension. In each case the surgical approach to the

perineum and caudal rectum was appropriate to the specific underlying pathology

present; thereafter either the internal obturator or semitendinosus muscle was elevated

in a standard fashion (Figure 1).

For internal obturator elevation, the attachment of the internal obturator muscle to the

ischium was first severed using a scalpel blade and a freer periosteal elevator was then

used to lift the muscle (and periosteum when possible) from the dorsal surface of the

ischium to the level of the obturator foramen. The tendon of insertion was cut to

mobilize the muscle laterally allowing the muscle to be reflected medially and to be

brought into apposition with the rectal wall. Polydioxanone (2M) was used to suture

the cranial, lateral and caudal margins of the internal obturator muscle to the

seromuscular layer of the rectum in a simple interrupted pattern (Hardie et al. 1983,

van Sluijs & Sjollema 1989, Aronson 2012).

For semitendinosus elevation, the surgical technique involved a proximo-distal skin

incision from the level of the ischiatic tuberosity extending down the caudal aspect of

the thigh to the level of the distal third of the hamstring muscles. The proximal aspect

of this skin incision was connected to the distal aspect of the skin incision created for

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

perineal / caudal rectum exploration to produce a bridging incision. The

semitendinosus muscle was dissected from the adjacent musculature and transected at

the appropriate level according to the length of muscle flap required to reach the rectal

defect (Figure 2). It was then rotated dorsally and sutured to the seromuscular layer of

the rectal wall using polydioxanone (1.5 or 2M) in a simple interrupted pattern. The

skin was closed routinely over the area of muscle dissection (Chambers & Rawlings

1991).

The three dogs sustained compromise of the rectal wall during excision of anal sac

adenocarcinoma neoplasms. The anal sac masses ranged from 30 to 43mm in

diameter; accidental iatrogenic rectal laceration occurred during mass dissection in

case 1, in case 2 no iatrogenic injury to the rectum was recognized during the initial

surgery, and intentional rectal wall excision was performed in case 3 due to

infiltration of the anal sac mass through the muscularis layer of the rectum. Immediate

primary appositional repair of the rectal wall tear was performed in case 1, using

1.5M polydioxanone in a single layer, simple continuous suture pattern and the

perineal soft tissues closed routinely. A pararectal abscess developed two days later,

exploration of which confirmed dehiscence of the rectal wall repair (Figure 3).

Subsequent attempts at primary appositional suture repair of the rectal wall were

unsuccessful and twelve days after the original surgery an internal obturator flap was

mobilised to successfully reinforce repeated suture repair of the rectal wall. This flap

was sutured to the dorsal aspect of the rectal wall (seromuscular layer) using 2M

polydioxanone in a simple interrupted pattern. In case 2 a pararectal abscess

developed two days after anal sacculectomy and, on exploration of the abscess pocket,

a rectal tear at the level of the ligated anal sac duct was confirmed (where the rectal

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

wall was thinnest following mass dissection). The rectal laceration was initially

repaired using 2M polydioxanone in a single layer, simple continuous suture pattern

and an internal obturator muscle flap used for reinforcement (Figure 4). The lateral

aspect of the muscle was reflected medially and sutured to the seromuscular layer of

the rectal wall, dorsal to the appositional repair, using 2M polydioxanone in a simple

interrupted pattern. Caudally the muscle flap was also sutured to the cranial edge of

the external anal sphincter muscle using the same suture material and pattern. Despite

the muscle flap on this occasion, recurrent pararectal abscessation was encountered

twenty-four hours later and a semitendinosus flap was subsequently employed. The

semitendinosus muscle was sutured over the re-apposed rectal wall (single layer,

simple interrupted sutures in 1.5M polydioxanone) and was anchored to the caudal

edge of the internal obturator flap and the coccygeus muscle. An active, closed-

suction, drain (Red-O-Pack, Vygon) was placed in the subcutaneous tissues prior to

closure of the skin and healing was achieved without further complications. In case 3

intentional excision of a portion of rectal wall was performed during tumour

dissection due to extension of the mass outside of the anal sac; the rectum was

repaired using 2M polydioxanone in a single layer, simple interrupted pattern and the

right internal obturator muscle was used successfully to reinforce the repair. Mild

faecal incontinence was apparent after surgery as a portion of the external anal

sphincter muscle was also excised due to tumour infiltration. This improved with time

but continued, intermittent tenesmus and occasional dropping of faeces was reported.

The cat (case 4) sustained a 1.5cm right-sided rectal wall laceration secondary to bite

wounds around her anus. Primary appositional repair of the rectum using 1.5M

polydioxanone in a single layer, simple interrupted suture pattern was performed.

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

Breakdown of the rectal wall repair occurred two days later and repeat single layer,

simple interrupted suture closure of the rectal wall was performed, again using 1.5M

polydioxanone. This appositional repair was reinforced by elevation of a right-sided

semitendinosus muscle flap which was anchored to the seromuscular layer of the

rectal wall around the site of the tear using a combination of simple continuous and

interrupted suture patterns in 2M polydioxanone. Routine skin closure was performed

with excision of the existing superficial wounds resulting in a ‘Y’ shaped incision and

healing progressed satisfactorily (Figure 5).

No peri- or postoperative complications related to internal obturator or

semitendinosus muscle elevation were described in the patient records reviewed or

during telephone follow up. Clinical examination and diagnostic findings,

descriptions of the surgical procedures performed and patient outcomes are

summarized in Table 1.

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

Discussion

The purpose of this report is to highlight the potential application of the internal

obturator and semitendinosus muscle transposition flaps to reinforce rectal wall repair

where primary appositional closure has failed, or may be expected to fail. In this case

series excisions of large anal sac neoplasms in three dogs and bite wounds involving

the rectum in one cat were the indications for muscle flap use, which avoided rectal

pull through and the unnecessary loss of functional rectal tissue.

In two of the three anal sac tumour excisions described (cases 1 and 3), rectal wall

laceration or intended rectal wall excision was recorded at the time of surgery; in the

third case (case 2) no apparent direct injury to the rectal wall was observed and

iatrogenic ischaemic damage was assumed. In every procedure described a bipolar

electrocautery device was applied judiciously for the purpose of haemostasis only.

Thermal devices were not used for mass dissection as such, rather a combination of

blunt and sharp dissection using small haemostats and Metzenbaum scissors was

chosen to avoid inadvertent damage to neurovascular structures and the muscles of the

perineal diaphragm where possible. It is recognised, however, that through the

application of electrocautery to small vessels supplying the anal sac masses either the

caudal rectal artery may have been damaged, or thermal necrosis of the rectal wall

may have been induced.

In two of the cases reported in this series (cases 1 and 4), single layer, appositional

closure of the rectum using either simple interrupted or simple continuous suture

patterns was attempted without adjunctive muscle flap mobilization in the first

instance. In both cases appositional suture closure was unsuccessful, despite this

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

being a recognized technique for primary repair of the rectal wall. On inspection the

suture material (polydioxanone in every case) was intact but rectal wall healing had

not progressed leading to re-opening of the original tear. The reason for failure of

healing is unknown, but is hypothesized to be in part due to the relatively poor blood

supply of the caudal rectum. In both cases there was a soft tissue cavity adjacent to

the rectal tear as a consequence of soft tissue loss by dissection of tumour or abscess

rupture, resulting in a lack of collateral blood supply from components of the pelvic

diaphragm compounding the limitations of the primary vascular network of the caudal

rectum. Both patients were receiving broad-spectrum antibiotics at the time of wound

breakdown, however concurrent infection remains a further possible contributing

factor to the failure of healing. Another factor that could play a role in the outcome of

primary rectal wall repair is the timing of the repair following injury – in some

circumstances it may be beneficial to delay repair until the tissues have had time to

declare themselves as necrotic or viable, but managing the resultant faecal

contamination of the perineal soft tissues in the meantime can be challenging.

Temporary rectal stenting (with or without negative pressure wound therapy) has been

described as a way of allowing short-term faecal diversion during perianal wound

management. However stent obstruction with faeces may be encountered, and the

morbidity associated with this technique requires further investigation (Skinner et al.

2016). In addition, it is hypothesised that delayed repair of the rectal tissue may lead

to fibrosis and retraction of the wall edges resulting in increased tension of a primary

appositional closure. Two-layer closure of the rectal wall was not attempted in this

series – premature suture absorption or suture failure was not observed in any of these

cases and, to the authors’ knowledge, there is no published evidence to suggest this

approach is preferable in colo-rectal surgery (Coolman et al. 2000, Aronson 2012).

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

In case 2, the appositional repair of the rectal wall broke down in spite of

reinforcement using an internal obturator flap. There was no obvious iatrogenic rectal

wall laceration during the tumour dissection in this case, and the subsequent rectal

wall breakdown was attributed to ischaemic necrosis. The second surgery, including

elevation of the internal obturator flap, occurred four days after the first, and it is

possible that the extent of the ischaemic damage was not grossly evident at this time

(Ellison 1989, Cornell 2012). The other possible explanation for failure of the internal

obturator flap to provide both structural and collateral vascular supply to the injured

rectal wall in this scenario is the anatomical limitations of this muscle flap. In the

authors’ experience, the internal obturator is restricted to supporting the lateral aspect

of the rectal wall only, and does not reach the more caudal rectum in most cases. In

the case of dorsal, ventral and very caudal rectal wall lesions, the semitendinosus

muscle transposition may therefore be more useful, and in case 2 addition of this flap

was seen to provide a definitive resolution.

An active, closed-suction drain was placed in case 2 following semitendinosus muscle

flap elevation. It is acknowledged that placement of such a drain may have been

beneficial in the management of the other cases to permit early detection of faecal

contamination of the tissues and minimize subsequent exposure of the tissues to faecal

material should dehiscence occur. The possible benefits of drain placement must

however be weighed against the potential disadvantages of ascending infection and

tumour seeding, and a decision made on a case-by-case basis.

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

Three of the cases included in this series underwent excision of anal sac

adenocarcinomas; in all cases metastatic spread of tumour to the sub-lumbar lymph

nodes was detected during staging. Whilst excision of the primary tumour, along with

the affected sub-lumbar lymph nodes, remains the treatment of choice in such a

scenario, it may not achieve disease cure and progressive lymphadenopathy or local

recurrence of tumour may be encountered. Chemotherapy and/or radiotherapy may be

considered as adjunctive treatments in the management of anal sac adenocarcinoma

disease, and various protocols have been described in the veterinary literature

(Bennett et al. 2002, Turek et al. 2003, Williams et al. 2003, Emms 2005). Whilst not

relevant for any of the cases described here, it is important to consider the potentially

greater radiation field required following mobilisation of a semitendinosus muscle

flap into a tumour bed, and the possible risks of this (ie skin burns and / or wound

dehiscence) for the patient.

In conclusion, it is the authors’ experience that primary appositional, single layer,

suture repair of lacerations affecting the caudal rectum is associated with a high risk

of dehiscence. In these situations application of either an internal obturator

transposition, and/or a semitendinosus muscle flap (particularly for more caudal or

ventral tears) has the potential to promote rectal wall healing, reducing the risk of

pararectal abscess formation. Further prospective, randomized studies are required to

verify these findings.

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

References

Aronson, L. R. (2012) Rectum, anus and perineum. In: Veterinary Surgery Small

Animal (Vol 2). Eds K. M. Tobias and S. A. Johnston. Elsevier Saunders, Missouri.

pp 1564-1600

Bennett, P. F., DeNicola, D. B., Bonney, P., Glickman, N. W. & Knapp, D. W. (2002)

Canine anal sac adenocarcinomas: clinical presentation and response to therapy.

Journal of Veterinary Internal Medicine 16 (1), 100-104

Bongartz, A., Carofiglio, F., Balligand, M., Heimann, M. & Hamaide, A. (2005) Use

of autogenous fascia lata graft for perineal herniorrhaphy in dogs. Veterinary Surgery

34(4), 405-413

Chambers, J. N., Rawlings, C. A. (1991) Applications of a semitendinosus muscle

flap in two dogs. Journal of the American Veterinary Medical Association 199 (1),

84-86

Coolman, B. R., Ehrhart, N., Pijanowski, G., et al. (2000) Comparison of skin staples

with sutures for anastomosis of the small intestine in dogs. Veterinary Surgery 29 (4),

293-302

Cornell, K. (2012) Wound healing. In Veterinary Surgery Small Animal (Vol 1). Eds

K. M. Tobias and S. A. Johnston. Elsevier Saunders, Missouri. pp 125-134

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

Doust, R. & Sullivan, M. (2003) Semitendinosus muscle transfer flap for external anal

sphincter incompetence in a dog. Journal of the American Veterinary Medical

Association 222 (10), 1385-1387

Ellison, G. W. (1989) Wound healing in the gastrointestinal tract. Seminars in

Veterinary Medicine and Surgery (Small Animal) 4 (4), 287-293

Emms, S. G. (2005) Anal sac tumours of the dog and their response to cytoreductive

surgery and chemotherapy. Australian Veterinary Journal 83 (6), 340-343

Evans, H. R. & de Lahunta, A. (2013) Abdominal Aorta. In: Miller’s Anatomy of the

Dog. 4th edn. Elsevier Saunders, Missouri. pp 498-499

Frankland, A. L. (1986) Use of porcine dermal collagen in the repair of perineal

hernia in dogs – a preliminary report. Veterinary Record 119 (1), 13-14

Goldsmid, S. E., Bellenger, C. R., Hopwood, P. R., et al. (1993) Colorectal blood

supply in dogs. American Journal of Veterinary Research 54 (11), 1948-53

Hall, J. C. Heel, K. A., Papadimitriou, J. M. & Platell, C. (1998) The pathobiology of

peritonitis. Gastroenterology 114 (1), 185-96

Hardie, E. M., Kolata, R. J. & Earley T. D. (1983) Evaluation of internal obturator

muscle transposition in treatment of perineal hernia in dogs. Veterinary Surgery 12,

69-72

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

Kirby B. M. (2012) Peritoneum and Retroperitoneum. In Veterinary Surgery Small

Animal (Vol 2). Eds K. M. Tobias and S. A. Johnston. Elsevier Saunders, Missouri.

pp 1398.

Morello, E., Martano, M., Squassino, C., et al. (2008) Transanal pull-through rectal

amputation for treatment of colorectal carcinoma in 11 dogs. Veterinary Surgery 37

(5), 420-426

Morello, E., Martano, M., Zabarino, S., et al (2015) Modified semitendinosus muscle

transposition to repair ventral perineal hernia in 14 dogs. Journal of Small Animal

Practice 56, 370-376

Mortari, A. C., Rahal, S. C., Resende, L. A. L., et al (2005) Electromyographical,

ultrasonographical and morphological modifications in semitendinosus muscle after

transposition as ventral perineal muscle flap. Journal of Veterinary Medicine. A,

Physiology, Pathology, Clinical Medicine 52, 359-365

Nucci, D. J., Liptak, J. M., Selmic, L. E., et al. (2014) Complications and outcomes

following rectal pull-through surgery in dogs with rectal masses: 74 cases (2000-

2013). Journal of the American Veterinary Medical Association 245 (6), 684-695

Orsher, R. J. (1986) Clinical and surgical parameters in dogs with perineal hernia.

Analysis of results with internal obturator transposition. Veterinary Surgery 15, 253-

258

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

Orsher, R. J. & Johnston, D. E. (1985) The surgical treatment of perineal hernia in

dogs by transposition of the obturator muscle. Compendium on Continuing Education

for the Practising Veterinarian 7, 233-239

Schallberger, S. P., Stanley, B. J., Hauptman, J. G. & Steficek, B. A. (2008) Effect of

porcine small intestinal submucosa on acute full-thickness wounds in dogs.

Veterinary Surgery 37 (6), 515-524

Shaughnessy, M. & Monnet, E. (2015) Internal obturator muscle transposition for

treatment of perineal hernia in dogs: 34 cases (1998-2012). Journal of the American

Veterinary Medical Association 246 (3), 321-326

Skinner, O. T., Cuddy, L. C., Coisman, J. G., Covey, J. L & Ellison, G. W. (2016)

Temporary rectal stenting for management of severe perineal wounds in two dogs.

Journal of the American Animal Hospital Association 52: doi 10.5326/JAAHA-MS-

6350 Epub ahead of print.

Stoll, M. R., Cook, J. L., Pope, E. R., Carson, W. L. & Kreeger, J. M. (2002) The use

of porcine small intestinal submucosa as a biomaterial for perineal herniorrhaphy in

the dog. Veterinary Surgery 31 (4), 379-390

Szabo, S., Wilkens, B., Radasch, R. M. (2007) Use of polypropylene mesh in

addition to internal obturator transposition: a review of 59 cases (2000-2004). Journal

of the American Animal Hospital Association 43, 136-142

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

Tobias, K. S. (1994) Rectal perforation, rectocutaneous fistula formation, and

enterocutaneous fistula formation after pelvic trauma in a dog. Journal of the

American Veterinary Medical Association 205(9),1292–1296

Turek, M. M., Forrest, L. J., Adams, W. M., Helfand, S. C. & Vail, D. M (2003)

Postoperative radiotherapy and mitoxantrone for anal sac adenocarcinoma in the dog:

15 cases (1991-2001). Veterinary and Comparative Oncology 1 (2), 94-104

van Sluijs, F. J. & Sjollema, B. E. (1989) Perineal hernia repair in the dog by

transposition of the internal obturator muscle. I. Surgical technique. Vet Quarterly

11(1), 12-17

Vnuk, D., Babic, T., Stejskal, D., et al. (2005) Application of a semitendinosus

muscle flap in the treatment of perineal hernia in a cat. Veterinary Record 156, 182-

184

Welches, C. D., Scavelli, T. D., Aronson, M. G., et al. (1992) Perineal hernia in the

cat: a retrospective study of 40 cases. Journal of the American Animal Hospital

Association 28, 431-438

Williams J. M. (2012) Colon. In Veterinary Surgery Small Animal (Vol 2). Eds K. M.

Tobias and S. A. Johnston. Elsevier Saunders, Missouri. pp 1548-1549

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

Williams, L. E., Gliatto, J. M., Dodge, R. K., Johnson, J. L., Gamblin, R. M., Thamm,

D. H., Lana, S. E., Szymkowski, M. & Moore, A. S. (2003) Carcinoma of the

apocrine glands of the anal sac in dogs: 113 cases (1985-1995). Journal of the

American Veterinary Medical Association 223 (6), 825-831

No conflicts of interest have been declared.

423

424

425

426

427

428

429

430